Google model product
Gemini 4 Argon
A Gemini 4 model from Google DeepMind for long-horizon software engineering, enterprise knowledge work, cybersecurity defense, and multimodal understanding of charts and video.
Updated Oct 23, 2026. Default version: Gemini 4 Argon
LLMBoard Score92.6Gemini 4 Argon
Coverage0%15 benchmark families
Context window1MTokens
Official input price$2Google API
Gemini 4 Argon Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Gemini 4 Argon Benchmark Results
Benchmark scores for Gemini 4 Argon.
Sort by
Benchmark | Score | Rank | Participants | Percentile | Evidence | Evaluated |
|---|
| BenchmarkCWE-Bench | Score68.00% | Rank01 | Participants2 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkGraphwalks BFS <128k | Score99.70% | Rank01 | Participants13 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkLABBench2 | Score88.80% | Rank01 | Participants3 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkLVBench | Score91.70% | Rank01 | Participants31 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkPostTrainBench | Score45.30% | Rank01 | Participants9 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkVals Finance Agent v2 | Score65.40% | Rank01 | Participants6 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkLM Arena Text | Score1,532.30 rating | Rank01 | Participants218 | Percentile100.00% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text Factuality | Score1,511.16 rating | Rank01 | Participants131 | Percentile100.00% | EvidenceA | Evaluated |
| BenchmarkLM Arena Text Style Control | Score1,523.93 rating | Rank01 | Participants218 | Percentile100.00% | EvidenceA | Evaluated |
| BenchmarkChartography (no tools) | Score71.60% | Rank01 | Participants3 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkRiemannBench | Score76.00% | Rank01 | Participants1 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkVals Index | Score68.90% | Rank01 | Participants1 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkVibe Code Bench | Score91.90% | Rank01 | Participants1 | Percentile100.00% | EvidenceC | Evaluated |
| BenchmarkGraphwalks BFS >128k | Score84.20% | Rank02 | Participants12 | Percentile90.91% | EvidenceC | Evaluated |
| BenchmarkHarvey's Legal Agent Benchmark | Score19.60% | Rank02 | Participants4 | Percentile66.67% | EvidenceC | Evaluated |
| BenchmarkLM Arena Agent Task Outcome Explicit | Score13.81% | Rank02 | Participants44 | Percentile97.67% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Leaderboard | Score12.03% | Rank03 | Participants44 | Percentile95.35% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Praise Complaint | Score27.31% | Rank03 | Participants44 | Percentile95.35% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Steerability | Score14.58% | Rank03 | Participants44 | Percentile95.35% | EvidenceA | Evaluated |
| BenchmarkFrontierSWE V2 | Score55.00% | Rank03 | Participants5 | Percentile50.00% | EvidenceC | Evaluated |
| BenchmarkTerminal-Bench-Science 0.1 | Score57.60% | Rank04 | Participants7 | Percentile50.00% | EvidenceC | Evaluated |
| BenchmarkDeepSWE 1.1 | Score77.90% | Rank05 | Participants51 | Percentile92.00% | EvidenceC | Evaluated |
| BenchmarkTerminal-Bench 4.0 | Score57.40% | Rank05 | Participants29 | Percentile85.71% | EvidenceC | Evaluated |
| BenchmarkLM Arena Webdev | Score1,678.26 rating | Rank05 | Participants107 | Percentile96.23% | EvidenceA | Evaluated |
| BenchmarkAutomationBench | Score51.30% | Rank06 | Participants27 | Percentile80.77% | EvidenceC | Evaluated |
| BenchmarkOSWorld 2.0 | Score69.20% | Rank06 | Participants17 | Percentile68.75% | EvidenceC | Evaluated |
| BenchmarkAgents' Last Exam | Score39.50% | Rank12 | Participants24 | Percentile52.17% | EvidenceC | Evaluated |
| BenchmarkLM Arena Agent Tool Hallucination | Score0.29% | Rank22 | Participants44 | Percentile51.16% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Bash Reliability No Prereq | Score-1.78% | Rank32 | Participants45 | Percentile29.55% | EvidenceA | Evaluated |
| BenchmarkLM Arena Agent Bash Reliability | Score-3.74% | Rank34 | Participants44 | Percentile23.26% | EvidenceA | Evaluated |
Gemini 4 Argon Arena Results
Preference and agent-evaluation results for the default version.
Sort by
Arena | Category | Rank | Rating / score | Votes | Observations | Result date |
|---|
| Arenatext | Categorychinese | Rank01 | Rating / score1,611.06 | Votes361 | ObservationsN/A | Result date |
| Arenatext | Categorycoding | Rank01 | Rating / score1,546.05 | Votes1,236 | ObservationsN/A | Result date |
| Arenatext | Categoryenglish | Rank01 | Rating / score1,537.62 | Votes1,900 | ObservationsN/A | Result date |
| Arenatext | Categoryexclude ties | Rank01 | Rating / score1,559.81 | Votes3,655 | ObservationsN/A | Result date |
| Arenatext | Categoryhard prompts | Rank01 | Rating / score1,546.00 | Votes3,131 | ObservationsN/A | Result date |
| Arenatext | Categoryhard prompts english | Rank01 | Rating / score1,547.58 | Votes1,208 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry business and management and financial operations | Rank01 | Rating / score1,524.59 | Votes890 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry entertainment and sports and media | Rank01 | Rating / score1,526.74 | Votes1,235 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry legal and government | Rank01 | Rating / score1,548.16 | Votes399 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry life and physical and social science | Rank01 | Rating / score1,547.91 | Votes818 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry mathematical | Rank01 | Rating / score1,557.49 | Votes331 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry medicine and healthcare | Rank01 | Rating / score1,554.76 | Votes332 | ObservationsN/A | Result date |
| Arenatext | Categoryindustry software and it services | Rank01 | Rating / score1,555.72 | Votes1,826 | ObservationsN/A | Result date |
| Arenatext | Categorylonger query | Rank01 | Rating / score1,546.82 | Votes2,063 | ObservationsN/A | Result date |
| Arenatext | Categorymath | Rank01 | Rating / score1,536.37 | Votes273 | ObservationsN/A | Result date |
| Arenatext | Categorymulti turn | Rank01 | Rating / score1,556.05 | Votes677 | ObservationsN/A | Result date |
| Arenatext | Categorynon english | Rank01 | Rating / score1,521.53 | Votes3,000 | ObservationsN/A | Result date |
| Arenatext | Categoryoverall | Rank01 | Rating / score1,532.30 | Votes4,900 | ObservationsN/A | Result date |
| Arenatext | Categoryrussian | Rank01 | Rating / score1,533.75 | Votes668 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryexclude ties | Rank01 | Rating / score1,535.15 | Votes3,655 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryindustry business and management and financial operations | Rank01 | Rating / score1,508.75 | Votes885 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryindustry entertainment and sports and media | Rank01 | Rating / score1,499.82 | Votes1,235 | ObservationsN/A | Result date |
| Arenatext factuality | Categorylonger query | Rank01 | Rating / score1,534.25 | Votes2,063 | ObservationsN/A | Result date |
| Arenatext factuality | Categorymulti turn | Rank01 | Rating / score1,526.01 | Votes671 | ObservationsN/A | Result date |
| Arenatext factuality | Categorynon english | Rank01 | Rating / score1,498.87 | Votes3,000 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryoverall | Rank01 | Rating / score1,511.16 | Votes4,900 | ObservationsN/A | Result date |
| Arenatext factuality | Categoryrussian | Rank01 | Rating / score1,518.43 | Votes651 | ObservationsN/A | Result date |
| Arenatext style control | Categorychinese | Rank01 | Rating / score1,588.95 | Votes361 | ObservationsN/A | Result date |
| Arenatext style control | Categorycoding | Rank01 | Rating / score1,558.70 | Votes1,236 | ObservationsN/A | Result date |
| Arenatext style control | Categoryenglish | Rank01 | Rating / score1,531.99 | Votes1,900 | ObservationsN/A | Result date |
Gemini 4 Argon Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- $2 input, $10 output per 1M
- Official provider
- Lowest third-party
- From $2 input, $10 output per 1M via NanoGPT
- Tracked offerings
- 23
Sort by
Provider | Provider model ID | Region | Input / 1M | Output / 1M | Context | Updated |
|---|
| ProviderNanoGPT | Provider model IDgoogle/gemini-4-argon | Regionglobal | Input / 1M$2 | Output / 1M$10 | Context1M | Updated |
| ProviderKilo Gateway | Provider model IDgoogle/gemini-4-argon | Regionglobal | Input / 1M$2 | Output / 1M$10 | Context1M | Updated |
| ProviderOpenRouter | Provider model IDgoogle/gemini-4-argon | Regionglobal | Input / 1M$2 | Output / 1M$10 | Context1M | Updated |
| ProviderGoogle | Provider model IDgemini-4-argon | Regionglobal | Input / 1M$2 | Output / 1M$10 | Context1M | Updated |
| ProviderVercel AI Gateway | Provider model IDgoogle/gemini-4-argon | Regionglobal | Input / 1M$2 | Output / 1M$10 | Context1M | Updated |
| ProviderMerge Gateway | Provider model IDgoogle/gemini-4-argon | Regionglobal | Input / 1M$2 | Output / 1M$10 | Context1M | Updated |
| ProviderOpenCode | Provider model IDgemini-4-argon | Regionglobal | Input / 1M$2 | Output / 1M$10 | Context1M | Updated |
| ProviderVertex | Provider model IDgemini-4-argon | Regionglobal | Input / 1M$2 | Output / 1M$10 | Context1M | Updated |
| ProviderAIHubMix | Provider model IDgemini-4-argon | Regionglobal | Input / 1M$2 | Output / 1M$10 | Context1M | Updated |
| ProviderGitHub Copilot | Provider model IDgemini-4-argon | Regionglobal | Input / 1M$2 | Output / 1M$10 | Context1M | Updated |
| ProviderRequesty | Provider model IDgemini-4-argon | Regionglobal | Input / 1M$2 | Output / 1M$10 | Context1M | Updated |
| ProviderCrossModel | Provider model IDgemini/gemini-4-argon | Regionglobal | Input / 1M$2 | Output / 1M$10 | Context1M | Updated |
| ProviderOfox | Provider model IDgoogle/gemini-4-argon | Regionglobal | Input / 1M$2 | Output / 1M$10 | Context1M | Updated |
| ProviderPrime Inference | Provider model IDgoogle/gemini-4-argon | Regionglobal | Input / 1M$2 | Output / 1M$10 | Context1M | Updated |
| ProviderCommandCode | Provider model IDgoogle/gemini-4-argon | Regionglobal | Input / 1M$2 | Output / 1M$10 | Context1M | Updated |
| ProviderCloudflare AI Gateway | Provider model IDgoogle/gemini-4-argon | Regionglobal | Input / 1M$2 | Output / 1M$10 | Context1M | Updated |
| ProviderAixy | Provider model IDgemini/gemini-4-argon | Regionglobal | Input / 1M$2 | Output / 1M$10 | Context1M | Updated |
| ProviderPerplexity Agent | Provider model IDgoogle/gemini-4-argon | Regionglobal | Input / 1M$2 | Output / 1M$10 | Context1M | Updated |
| ProviderCortecs | Provider model IDgemini-4-argon | Regionglobal | Input / 1M$2.25 | Output / 1M$11.25 | Context1M | Updated |
| ProviderVenice AI | Provider model IDgemini-4-argon | Regionglobal | Input / 1M$2.5 | Output / 1M$12.5 | Context1M | Updated |
| ProviderEden AI | Provider model IDgoogle/gemini-4-argon | Regionglobal | Input / 1M$4 | Output / 1M$20 | Context1M | Updated |
| ProviderDevPass (LLM Gateway) | Provider model IDgemini-4-argon | Regionglobal | Input / 1M$4 | Output / 1M$20 | Context1M | Updated |
| Provider302.AI | Provider model IDgemini-4-argon | Regionglobal | Input / 1M$4 | Output / 1M$10 | Context1M | Updated |
Gemini 4 Argon Runtime Performance
Provider-specific output speed and catalog latency for Gemini 4 Argon. Runtime does not affect the capability score.
No runtime data
No provider-specific speed or latency record is linked to the default version yet.
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Gemini 4 Argon Specifications
Technical details for the model's default version.
- Version
- Gemini 4 Argon
- Released
- Sep 30, 2026
- Knowledge cutoff
- Unknown
- Parameters
- N/A
- Context window
- 1M
- Max output
- 1M
- Inputs
- audio, image, text, video
- Outputs
- text
- Open weights
- No
- License
- Proprietary
Gemini 4 Argon Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
Sort by
Version | Released | LLMBoard | Parameters | Context | Max output | Open weights | License |
|---|
| VersionGemini 4 Argon | Released | LLMBoard92.57 | ParametersN/A | Context1M | Max output1M | Open weightsNo | LicenseProprietary |
Models similar to Gemini 4 Argon
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
GO
Google
Gemini 3.8 Flash
GOGoogle
Gemini 3.7 Flash
GOGoogle
Gemini 3.1 Pro
GOGoogle
Gemini 3.5 Flash
GOGoogle
Gemini 3.6 Flash
GOGoogle
Gemini 3 Pro
What is Gemini 4 Argon?
Key information about Gemini 4 Argon and its available data.
Gemini 4 Argon is a Google DeepMind large language model with a 1M-token output limit, up from 64K. As of early October 2026, it is rolling out only to trusted cyber defenders through Google's Fairwind Program, with broader access planned for paid API customers and Google AI Ultra subscribers.
Data as of 2026-10-22.
FAQ
Common questions about Gemini 4 Argon.
When was Gemini 4 Argon released?
Gemini 4 Argon's default version was released on Sep 30, 2026.
How much does Gemini 4 Argon cost?
Gemini 4 Argon's official API price is $2 per million input tokens and $10 per million output tokens via Google. The lowest tracked third-party offer starts at $2 input and $10 output via NanoGPT.
Who created Gemini 4 Argon?
Gemini 4 Argon was created by Google.
What is the context window for Gemini 4 Argon?
The default version has a 1M token context window.
Is Gemini 4 Argon open weight?
No. The default version is not marked as having publicly available weights.